Reconstruction of a tibial defect with microvascular transfer of a previously fractured fibula.

A 43-year-old man sustained severe injuries to his lower limbs with extensive soft-tissue damage and bilateral tibial-fibular fractures. Acutely, the patient underwent external fixation and a free latissimus dorsi flap for soft-tissue coverage of the left leg. However, the tibia had a nonviable butterfly fragment that left a 7-cm defect after debridement. Subsequently, the contralateral fractured fibula was used as a bridging vascularized graft for this tibial defect. The transfer of a fibula containing the zone of injury from a previous high-energy fracture has not been reported. This case demonstrates the successful microvascular transfer of a previously fractured fibula for the repair of a contralateral tibial bony defect.

Humorally mediated thrombocytosis in major lower extremity trauma.

Thrombocytosis in patients undergoing free tissue transfer for coverage of posttraumatic lower extremity defects may be associated with an increased incidence of microvascular thrombosis. Patients with isolated lower extremity trauma have an elevated platelet count that peaks approximately 2 weeks after injury. It is our theory that a humoral component of trauma sera is responsible for the induction of this thrombocytosis. Eight patients with isolated soft-tissue and bony trauma were included in the study. Serum was collected at baseline and throughout the study period. Platelet count, leukocyte count, hemoglobin concentration, and hematocrit were determined. Immunoassay for human interleukin-3 (IL-3), IL-6, and IL-11 as well as granulocyte macrophage colony stimulating factor (GM-CSF) were performed by solid-phase enzyme-linked immunosorbent assay. Balb-C mice were then injected intraperitoneally with the human trauma sera from all time points. Blood was collected at baseline and throughout the study period for determination of platelet count, hemoglobin, and hematocrit. Mean initial platelet count in the 8 human subjects was 152,000 per cubic millimeter with an average peak to 642,000 per cubic millimeter. IL-3, IL-11, and GM-CSF were not detectable in the serum of any patient. Elevated levels of IL-6 were detected in all patients in a nonspecific pattern. In the murine model, an early and late thrombocytosis was elicited. The early peak averaged 78.6% over baseline whereas the late peak average 81.0% over baseline. The induction by human trauma sera of an early and late thrombocytosis in this mouse bioassay supports the theory of humoral mediators. The humoral mediators are yet to be determined but may include IL-6.

A nerve distraction model in the rat.

Segmental loss of a peripheral nerve has been a challenging reconstructive problem. Management of the nerve gap has been accomplished classically with nerve grafting. However, autogenous nerve grafts are not always available for bridging large nerve gaps, and clinical results of large nerve cable grafts have been disappointing. Newer techniques concentrate on nerve lengthening with different methods. Tissue expansion of peripheral nerves has been producing promising results. Since the introduction of the Ilizarov external fixator, much attention has turned to limb-lengthening techniques and studies investigating the results of nerve and soft tissues lengthened during the course of this procedure. Primary nerve distraction may be an alternative to nerve elongation, by expansion or nerve grafting to repair the peripheral nerve gap. This study describes a device and a model for peripheral nerve distraction in a rat. Primary nerve distraction will need to be subjected to vigorous studies before clinical application.

In vitro prefabrication of human cartilage shapes using fibrin glue and human chondrocytes.

We report the first generation of human cartilage from fibrin glue using a technique of molding chondrocytes in fibrin glue developed in our laboratory. Human costal chondrocytes were suspended in cryoprecipitate and polymerized into a human nasal shape with bovine thrombin. After culture in vitro for 4 weeks, this construct was implanted subcutaneously into a nude mouse. The final construct harvested after 4 weeks in vivo demonstrated some preservation of its original features. Histological analysis showed features of native cartilage, including matrix synthesis and viable chondrocytes by nuclear staining. Biochemical analysis demonstrated active matrix production. Biomechanical testing was performed. To our knowledge this is the first reported creation of human cartilage from fibrin glue, and the first creation of human cartilage in vitro. This technique may become a promising means of engineering precisely designed autogenous cartilage for human reconstruction.

Balloon assisted endoscopic harvest of the latissimus dorsi muscle.

In this study, we present our experience with balloon assisted endoscopic harvest of the latissimus dorsi muscle for extremity reconstruction. The balloon performs most of the dissection under the muscle and creates the optical work space used in the endoscopic dissection. Over the course of this series the operative time has been reduced and averaged 2 hours and 44 minutes. The reconstructive goals were met in all cases. The average axillary incision length was 5.6 cm, and there were an average of 1.3 one-centimeter or smaller counter incisions.

Venous microanastomosis with the Unilink system, sleeve, and suture techniques: a comparative study in the rat.

Multiple microvascular anastomotic techniques have been described with varying rates of success. This experimental study presents the results of a comparison of three types of venous microanastomotic techniques: the Unilink system, the sleeve technique, and the suture technique. Twenty male Sprague-Dawley rats, 40 femoral veins, were used for this study. In vivo observation and microvasculography demonstrated that patency rates between the Unilink system and suture techniques were comparable (p > 0.05) and were significantly superior to the sleeve anastomosis (p < 0.05). The anastomotic time for the sleeve technique was significantly shorter than for the suture technique (p < 0.001). Compared with suture and sleeve anastomoses, the anastomotic time employing the Unilink system was significantly the shortest (p < 0.001). The Unilink system proved to be the fastest method with the highest patency rate. These results suggest that the use of the Unilink system is superior with regard to anastomotic time and patency rate, when compared to suture and sleeve techniques for venous microanastomosis.

Selective use of preoperative lower extremity arteriography in free flap reconstruction.

Preoperative angiography is commonly utilized prior to free flap reconstruction of the lower extremity. The charts and radiographs of 38 patients who underwent free flap reconstruction, after acute posttraumatic lower extremity injuries, were studied. Patients were categorized according to the presence or absence of vascular abnormality based on pulse examination alone. Specific vascular abnormalities were recorded in each group. Of the 38 patients who had preoperative lower extremity arteriography, 23 were found to have normal dorsalis pedis and posterior tibial pulses. Only 1 of these patients had an angiographic abnormality. Of the 15 patients with abnormal pulse examinations, all were found to have angiographic abnormalities. Cost analysis of the lower extremity angiogram revealed a total additional expense of $2,957. Pulse examination was found to be a sensitive and effective predictor of lower extremity vascular integrity. Although lower extremity angiography is encouraged when distal pulse examination is abnormal, the use of preoperative arteriography for lower extremity microvascular free flap reconstruction is probably unnecessary in most patients with normal distal pulses.

Axonal regeneration through an autogenous nerve bypass: an experimental study in the rat.

Neuroma-in-continuity can manifest itself not only as pain but also as incomplete return of motor and sensory nerve function. The mainstay of current treatment for peripheral neuromas employs neurolysis or segmental resection with interposition grafting. These techniques are complicated by the loss of the remaining conduction through intact fibers within the injured segment. Based on the recent finding that end-to-side neurorrhaphies demonstrate axonal growth, we studied the use of a nerve "bypass" graft as a possible alternative to neurolysis or segmental resection with interposition grafting. A sciatic nerve crush injury model was induced in the Sprague-Dawley rat by compression with a straight hemostat. Epineurial windows were created proximal and distal to the injury. An 8-mm segment of radial nerve was harvested and anastomosed to the sciatic nerve at the epineurial window sites proximal and distal to the compressed segment (bypass group). A sciatic nerve crush injury without bypass served as a control. Electrophysiological testing and gate studies were performed over an 8-week period. Sciatic nerves were then harvested en bloc and studied under transmission electron microscopy at 1250 times magnification. Myelinated and unmyelinated axon counts were obtained. Nerve conduction velocity in the bypass group was significantly faster than conduction velocity in the control group at 8 weeks (44.8 m per second vs. 36.4 m per second; p = 0.031). We found no difference in myelinated axon counts between the proximal and distal segments of the control sciatic nerve. In the experimental sciatic nerve, a 160% increase in the number of myelinated axons was noted in the distal segment. Significant axonal growth was noted in the bypass nerve segment itself. Gait analysis using the sciatic functional index revealed improved function of the bypass group compared to the control group, but this was not statistically significant. Nerve bypass may serve to augment peripheral axonal growth while avoiding further loss of the native nerve.

Fate of free flap microanastomosis distal to the zone of injury in lower extremity trauma.

The decision to perform free flap microanastomosis to clearly uninjured vessels proximal to the zone of injury for lower extremity reconstruction must be weighed against the anatomic and technical difficulties of performing such an anastomosis. Preserved blood flow through vessels traversing the zone of injury has been shown. The records of all patients who underwent lower extremity reconstruction with microvascular free flaps at NYU Medical Center and Bellevue Hospital Center from January 1979 through August 1995 were reviewed. Patients with free flap microanastomoses distal to the zone of injury were compared with those with proximally based anastomoses. The group of patients was subdivided further into acute (1-21 days), subacute (22-60 days), and chronic (greater than 60 days) reconstruction groups. Of 451 microvascular free flaps, 35 were performed with recipient vessels distal to the zone of injury. Time interval from injury to coverage ranged from 24 hours to 57 years. Of 35 distally based flaps, 33 (94 percent) were successful and 5 required reoperation (14 percent). There was a similar incidence of thrombotic complications throughout all after-injury phases. Of 416 free flaps performed with microanastomoses to vessels proximal to the zone of injury, 388 (93 percent) were successful and 62 (15 percent) required reoperation. There was no significant difference (p > 0.05) in outcome between distal and proximal anastomoses and no significant difference (p > 0.05) in rates of reoperation. Timing of operation after injury had no bearing on outcome. Distally based microvascular free flaps anastomoses may be technically less difficult with rates of survival equal to those of proximally based flaps. The consideration and use of microanastomoses distal to the zone of injury are encouraged in selected patients.

The fate of lower extremities with failed free flaps.

This study reviews the outcome of patients with failed free flaps to lower extremities. The failure rate was 10 percent (41 of 413 flaps) over a 13-year period. Trauma patients (83 percent of all patients) had a failure rate of 11 percent, while nontrauma patients had a failure rate of 6.7 percent. The most common cause of failure was venous thrombosis (34 percent). Eight of 36 patients (22 percent) went on to amputation after the failed free flap; all were trauma patients. Patients with tibia-fibula fractures had a 35 percent amputation rate (6 of 17 patients) after a failed free flap. Seventy-eight percent of the patients (28 of 36) had salvage of their extremities by split-thickness skin graft, local flaps, or a second free flap. Long-term follow-up was available in 24 of 36 patients (67 percent), 20 of whom were salvaged without amputation. Of the patients whose limbs were salvaged, none had undergone an amputation at a mean follow-up of 6.2 years. All were ambulating, but 7 (35 percent) had intermittent wound breakdown. Despite an initial free-flap loss, the majority of extremities can be salvaged with subsequent procedures. However, on long-term follow-up, a large percentage of patients continue to have wound problems.

Thrombocytosis after major lower extremity trauma: mechanism and possible role in free flap failure.

Microvascular thrombosis and free flap failure are complications of free tissue transfer for coverage of lower extremity soft-tissue and bony defects despite appropriate vessel selection and adherence to meticulous technique. Increased rates of flap failure have been associated with reconstruction performed between 3 days and 6 weeks after injury, as well as in patients with thrombocytosis. We have found that serum platelet levels rise significantly after lower extremity injury. It is our theory that a circulating mediator or cytokine is released in response to injury, inducing the thrombocytosis. Twenty-one patients with Gustilo grade IIIb and IIIc injuries were studied prospectively. Serum was collected throughout the postinjury period. Platelet count, leukocyte count, hemoglobin concentration, and hematocrit were determined. Samples were also subjected to a platelet aggregation study as well as enzyme-linked immunosorbent assay for interleukin-3, interleukin-6, interleukin-11, and granulocyte macrophage-colony-stimulating factor. Megakaryocyte growth and development factor enzyme-linked immunosorbent assay and a myleoproliferative leukemia virus-transfected cell line assay for thrombopoietin were performed. Bone marrow was studied with flow cytometric analysis. Mean initial platelet count was 196,000 per cubic millimeter. There was an initial 26% decline to 140,000 per cubic millimeter, followed by an increase to 361% of baseline on day 16. No significant variations in serum leukocyte count or hemoglobin concentration were seen. Spontaneous and induced platelet aggregation responses were normal. Interleukin-6 was detected at elevated levels. However, interleukin-3, interleukin-11, granulocyte macrophage-colony-stimulating factor, and thrombopoietin were not measurable. Marked megakaryocytosis was seen on bone marrow analysis. Interleukin-6 may, therefore, play a role in the mechanism of thrombocytosis. We suggest that because patients with complex bony injuries of the leg experience platelet elevations that peak approximately 2 weeks after injury, microvascular free flap reconstructions should be considered high risk during this time period.

Microsurgical reconstruction of the lower extremity using the 3M microvascular coupling device in venous anastomoses.

Microsurgical reconstruction of the lower extremity presents a difficult problem to plastic surgeons; the rate of failure is higher than any other anatomical site. We reviewed our recent experience with lower extremity microsurgical reconstruction using the 3M vascular coupling device. We believe the excellent patency rate of the coupler may minimize the well-described problem of venous thrombosis in this challenging group of patients. This study involves a consecutive series of 11 patients who presented for reconstructive microsurgery of the lower extremity at NYU Medical Center hospitals between June 1 and September 1, 1994. Ten of 11 patients had free flap transfer to traumatic lower extremity injuries, whereas the remaining reconstruction was in a diabetic individual with a chronic wound. Fifteen microvascular venous anastomoses were performed; all but 1 was performed using the 3M coupler. Our experience with 11 patients, involving 14 mechanically coupled venous anastomoses, demonstrated successful use of the coupler. No intraoperative or postoperative vascular complications occurred. The overall success rate of the 3M coupler for venous anastomoses was 100%, and all microvascular free flaps were successful. We recommend using the 3M coupling device for venous anastomoses during reconstructive microsurgery of the lower extremity. Our series demonstrates the safety and effectiveness of the 3M coupler in this challenging group of patients. In addition, a secondary benefit of the 3M coupler is a significant reduction in operative time.

Salvage of traumatic below-knee amputation stumps utilizing the filet of foot free flap: critical evaluation of six cases.

Over a 12-year period between 1979 and 1991, 27 patients were operated on at the New York University Medical Center for salvage of below-knee amputation stumps utilizing free flaps. Six different donor sites were used. In 6 patients, the amputated foot was the donor site for a free flap to cover the tibial stump. There were 3 males and 3 females in this group. Five of the patients underwent immediate filet of foot reconstructions, while 1 patient had a reconstruction performed 69 days after injury, electively, when it was determined that below-knee amputation was the best option. All foot flaps survived and ultimately provided the major soft-tissue coverage for the below-knee amputation stump. The length of hospitalization ranged from 24 to 118 days. The time required from foot filet procedure to ambulation was 2, 4, 6, 7, 9, and 12 months in the 6 patients. Five of the 6 patients have resumed work or school after their injury. Foot flaps were based on the posterior tibial artery, anterior tibial artery, or both vessels. Nerve anastomosis of the posterior tibial nerve was performed in 5 patients. In 1 patient it was possible to maintain the continuity of the posterior tibial nerve. Five of the 6 patients were tested over a year after the flap, and all have good cold, pressure, and vibration sensation. Two of the 5 patients have heat sensation, and all 5 patients have at least protective pressure sensation. All the patients ambulate well with a below-knee prosthesis.

Limb salvage with microvascular free flap reconstruction using simultaneous polytetrafluoroethylene graft for inflow.

Microvascular free flaps have been successfully used to cover defects of the lower extremity. In patients with peripheral vascular disease and lower extremity defects, revascularization with in situ or reversed saphenous vein bypass graft combined with microvascular tissue transfer can salvage a limb that would otherwise be amputated. However, some of these patients may not have autologous vein available for the bypass procedure. We present a case of a 69-year-old man who underwent revascularization with a long polytetrafluoroethylene (PTFE) graft and a simultaneous microvascular free flap reconstruction using the PTFE graft as the inflow. The patient had undergone coronary artery bypass graft with saphenous vein and experienced a nonhealing wound of the distal saphenous vein harvest site and exposure of 8 cm of tibia. Angiogram revealed a significant stenosis of the common iliac artery, occluded superficial femoral artery, faint filling of the profunda femoris artery, and a faintly reconstituted posterior tibial artery. Because the patient had no available saphenous vein for bypass, he underwent an axillary-profunda and profunda-posterior tibial artery bypass with PTFE. A rectus abdominus microvascular free flap with direct anastomosis of the inferior epigastric artery to the PTFE was used to cover the exposed bone. The patient currently ambulates without difficulty. Limb salvage using bypass with PTFE combined with simultaneous microvascular free flap reconstruction is possible in selected patients.

Endoscopic harvest of the rectus abdominis free flap: balloon dissection in the fascial plane.

Free-flap donor sites are a frequent source of morbidity, including scar deformity and reduced mobility, as well as a significant contributor to recovery time after surgery. We present our technique for endoscopic harvest of the rectus abdominis muscle. A groin crease incision is made, which allows identification of the vascular pedicle and access to the inferior portion of the muscle. A balloon dissection device is inserted along the posterior rectus sheath and inflated. The inferior incision is closed over an endoscopic port after medial and lateral ports are inserted under direct vision at the level of the umbilicus. The cavity is insufflated with carbon dioxide, allowing visualization using a 10-mm, 30-degree endoscope. The remaining dissection is performed sharply, and the muscle is harvested via the groin incision. This technique has proved feasible during study in fresh human cadavers. Insufflation greatly reduces work load with retractors. The balloon device speeds dissection with a minimum of trauma. Because all dissection is performed from within the rectus sheath, the peritoneal cavity is not violated. Endoscopic rectus abdominis harvest using the fascial plane is safe and efficient and carries the potential to reduce donor-site morbidity.

Microvascular free-flap salvage of the diabetic foot: a 5-year experience.

This study reviews 21 microvascular free flaps to the diabetic foot in 19 patients over a 65-month period. All flaps were either to the plantar surface of the foot or to cover exposed Achilles tendon. Twenty of the flaps survived. The operations required a long, costly hospitalization with frequent recipient- and donor-site complications. All patients eventually ambulated on their flaps. Five patients came to proximal amputation from 6 to 37 months after surgery. Only one amputation was for flap breakdown.

The role of microvascular free flaps in salvaging below-knee amputation stumps: a review of 22 cases.

Twenty-two cases of traumatic below-knee amputation stumps with inadequate soft-tissue coverage salvaged with microvascular free flaps were reviewed retrospectively. All patients would have required an above-knee amputation for prosthesis fitting had microvascular free flaps not bee utilized. A total of 24 flaps were used in 22 patients; parascapular 11 (46%), foot filet six (25%), latissimus dorsi four (17%), lateral thigh, tensor fascia lata, and groin one (4%). Free flaps were performed immediately after injury in five (21%) cases, within the first week in two (8%), between 1 and 3 months in 12 (50%), and after 3 months in five (21%). Fifty per cent of the patients had significant other injuries. The patients had a total of 107 operations (mean, 4.9) related to their injury: 33 (mean, 1.5) of those operations were after the free flap, 27 (25%) of which were either performed because of a complication of the free flap or for revision of the free flap. Complications included partial necrosis in five (21%), neuroma in three (13%), hematoma in two (8%), donor site complication in two (8%), thrombosis requiring reoperation in one (4%), and flap failure in one (4%). Patient followup ranged from 12 to 116 months. All patients maintained a functional below-knee prosthetic level. The mean time to ambulation was 5.75 months, and was not significantly affected by flap complications. Most patients employed before their injury were employed after their injury. Despite a protracted course in these severe injured trauma patients, a functional below-knee amputation level was preserved in all cases utilizing microvascular free flaps.